Smart Metro: Deep Learning Approaches to Forecasting the MRT Line 3 Ridership

Abstract

Purpose – Since its establishment in 1999, the Metro Rail Transit Line 3 (MRT3) has served as a transportation option for numerous passengers in Metro Manila, Philippines. The Philippine government's transportation department records more than a thousand people using the MRT3 daily and forecasting the daily passenger count may be rather challenging. The MRT3's daily ridership fluctuates owing to variables such as holidays, working days, and other unexpected issues. Commuters do not know how many other commuters are on their route on a given day, which may hinder their ability to plan an efficient itinerary. Currently, the DOTr depends on spreadsheets containing historical data, which might be challenging to examine. This study presents a time series prediction of daily traffic to anticipate future attendance at a particular station on specific days.

Method – The proposed prediction approach uses DOTr ridership data to train multiple models that can provide correct data on Azure AutoML. These trained models have the highest accuracy: Gradient Boosting, Extreme Random Trees, and Light GBM.

Results – Based on historical data, this study aims to build and evaluate several prediction models for estimating the number of riders per station. On Azure AutoML, the Gradient Boosting, Extreme Random Trees, and Light GBM algorithms were investigated and executed. Gradient Boosting and Extreme Random Trees frequently made the most accurate predictions of the three algorithms, with an average accuracy of over 90%.

Conclusion – This research aims to develop and test different models of prediction for forecasting the number of riders per station based on historical data. Seven days of data were utilized for applying the model or assessing its correctness. Each model's resultant accuracy in each station is unique and may be modified by ridership and geography. However, the model still provides complete precision. Accuracy may be enhanced if additional current, valuable, and efficient characteristics are introduced to the dataset. MRT3 might incorporate a mortality rate component into the station's relative location or passenger capacity.

Recommendation – As the acquired data were from a pandemic, it is suggested that additional information be employed in future research. The circumstances of the MRT might change substantially over time; therefore, it is essential to refresh the training dataset.

Practical Implication – There are several benefits to applying time series forecasting in predicting the ridership of the MRT3 in the Philippines. This can allow decision-makers to make informed decisions about optimizing the MRT3 system to meet the needs of commuters. Additionally, time series forecasting can help to identify potential problems or issues in advance, such as overcrowding or maintenance needs, allowing for proactive solutions to be implemented.